I'm wondering how accurate and dependable the Phantom 3 barameter is.  In several occassions, I'm noticed a tendancy to drift.  For instance, on my last flight I hovered about 10 feet above a large tree.  To my surprise a few minutes later, my Phantom was almost touching the leaves of the same tree.  When I landed at the original takeoff point, my indicated altitude was still 60 feet.  Can I trust my Phantom to hold a height above an object without slowly descending into it?  Why is my indicated altitude almost 0 at takeoff and 60 feet when I land a the same location after a normal flight?  Has anyone else had this problem.  Is there some solution?

Man, finally that is exactly the same thing mine is doing, exactly!  I have posted at least a hundred times on this problem stating exactly what you just said,  I think lots of other are reporting the problem in different ways, and not being specific,  We are not alone.

I keep flipping back and forth between hope and despair that a software fix will cure our problem, I believe it will,  It has been incredibly difficult to get  any reasonably intelligent  responses to this  issue at all, look around through the threads, and you will see what I mean.

A guy named Mark was accused as being a drama queen when he pointed out, that this issue may be dangerous,  I was accused of smoking to good a dope, when I expressed my concerns as to safety,
Others have been attacked over there feelings of possible dangers involved with this issue.

We need to gather some more specific information on this situation, of this seeming altitude drift, It is of concern to some and perceived as  dangerous to others. There has been some serious fly offs with people getting hurt, property damaged and phantoms destroyed.

I urge those with this or other altitude, barometric sensor questions, and safety concerns with the telemetry errors,  post your information on this thread, We may be able to get this problem recognized  as the real problem,  that it is .  It may be relative to larger issues,  I ask those that say it is not ,  How do you know?  Are you developers? software engineers maybe?

Lets gather up some real, and specific, intelligently based  info so as we may help DJI identify and fix some of these issues, that is creating all of this "drama" , so as we can get on to feeling confident and safe flying this incredible machine. thanks Bob

mtnmaddman Posted at 2015-6-22 13:42
Man, finally that is exactly the same thing mine is doing, exactly!  I have posted at least a hundre ...

The last few flights I've had issues.  It seems like if it isn't rock steady when I first lift off, then it gets very difficult to control if I try to fly it.  Crashed into the ground just after lift off twice. It rises two or three feet then starts swaying/swinging wildly from side to side/front to back.  It's very difficult to bring back down without crashing.  

mtnmaddman Posted at 2015-6-22 13:42
Man, finally that is exactly the same thing mine is doing, exactly!  I have posted at least a hundre ...

Me also dji support asked for my flight logs. Updated to 1.2.6 calibrated imu, calibrated compass took it to a local empty field and straight away after take off drifting in circles changing altitude all on its own. I haven't flown it since 15/06/15. To nervous. Im pretty sure Splodge will be around soon telling us it's operator error, and dji are a lovely helpful bunch of people. Im a electronics engineer so let's say I know what I'm doing, and for the moment my phantom 3 is as reliable as a chocolate teapot. Perhaps the statements from frank Wang saying the phantom 3 was not a complete product is starting to make sense.

Philliplloyd20 Posted at 2015-6-22 14:21
Me also dji support asked for my flight logs. Updated to 1.2.6 calibrated imu, calibrated compass  ...

Have whatever opinion you want Phillip. I am only trying to help people but clearly you don't need help and have it all angles covered with the Phantom on a shelf sitting in a corner too nervous to fly it.

I have never told people its user error ( some situations seem to be though ), if anything I'm generally more honest about what i think the issues are than most.

You cannot hide form the fact that MOST are NOT having issues. So its not a case of the P3 is bad or the firmware is dodgy, its a case of finding why its affecting a tiny minority of users. The only way to do this is either the user sends it back to DJI or provides info on a forum for others to TRY and help. Moaning throwing crazy accusations about does not solve it neither does giving people who try and help grief..

Also, in my line of work, when I hear statements like "I'm a electronics engineer so lets say I know what I'm doing" usually means they don't!

Enjoy the chocolate teapot whilst I plan of having a fly at lunchtime

Is this with the latest firmware Bob? People have reported some crazy altitude readings with it and also the VPS is causing stability issues when its within range. For now people with issues maintaining altitude and position are advised to turn off VPS. Although some still repotr the issue..

Both issues have been reported to DJI but I don't have an ETA for a fix. I have tried to re-create this issue but apart from a foot or so difference could not see the barometer issue. I only have one Phantom 3 pro at the moment to test with as my other one went to the DJI never-land with jello/gimbal issues . I do have a P3P on order so I can get more flying and also a Phantom I can compare firmware versions with in future. This will probably make it easier to compare when there are issues like this.

I've noticed the same thing even with firmware 1.1.9 on my Phantom 3 Pro.  At the moment I'm super careful about taking it too close to the ground to get a shot.  I'm hoping it's something that can be fixed in a firmware update rather than being a hardware issue.  

sploodge Posted at 2015-6-22 15:26
Have whatever opinion you want Phillip. I am only trying to help people but clearly you don't need ...

Lead bosch software developer for numerous car manufacturers. Abs ecus pcm modules anything that requires software to function correctly. Im not taking any chances flying until I'm personally happy that the latest release for the p3p is stable. Not to worry tho because I will use my spare p3p that is on 1.1.9..
Unfortunately my  phantom 3 was affected by the 1.2.6 firmware update .now please go and do some work I'm sure your employer pays you to work not constantly reply to nearly every single post on this forum. One last thing, liking DJI'S arse is not going to get you a job with them! Would you like confirmation of my status working for bosch I'm more than happy to point you in a direction for your own clarification.  

sploodge Posted at 2015-6-22 15:36
Is this with the latest firmware Bob? People have reported some crazy altitude readings with it and  ...

Sploodge,
  Determining root cause of an issue is more difficult than just purchasing two or three of the same item and installing different software to see if you can re-create an issue.

  Electronic components are purchased with tolerances from a multitude of suppliers and have variances allowed per specification, Now enter in the human element of assembly and you have a hodgepodge of parts all combining for something that works and something that doesn't.

  Read up on Six Sigma. And for that matter, Just in Time Inventory. Is DJI performing all of the electronics assembly themselves or have the sub contracted it out to a single or multiple sources. The same goes for final assembly. What in-process QC checks are performed and what do they do when there is a deviation?

  Your first Phantom 3 Pro in most cases won't have the same components by the same supplier with the same values as Phantom #250,000.

  The Phantom issues are very complex and require a full analysis using sound engineering tools. Pareto analysis needs to be performed and once the data set is entered, will point you in the direction of the solution.

SQ# of Phantom
Pilot App version
Configuration of Pilot app
OS rev #
FPV used
Installed programs
Programs in use
Google Play store version if Android
Calibration of IMU
Calibration of compass
Ambient temperature
Users elevation
Number of satellites
Problems observed

Well, you get the idea.
  

gregg1r Posted at 2015-6-22 20:33
Sploodge,
  Determining root cause of an issue is more difficult than just purchasing two or thre ...

"Gee, Mr. Doe. We're sure sorry about your Phantom that ran amok, seriously injuring several people. I hope you're aware that *most* of our customers have not experienced such problems."

Of course, if such an issue ended up in court, it isn't very likely to end up in a court in you-know-where. And if the going got really, really tough, there's always the technique employed a number of years ago by Aero-Spatiale (sp?). They are a French firm manufacturing helicopters (and other products). After a life-flight crash in the Kansas City area, a huge lawsuit began. The French firm said goodbye and stopped selling that product in the U.S. End of discussion/suit.

The U.S. is either the only, or one of very few, countries that allow product liability lawsuits.

Personally speaking, does that mean I think my Phantom 3 is dangerous? No, but when I fly it now, I'm very, very careful and I haven't done some of the things I *might* have done before a number of these problems (I too seem to be experiencing some sort of VPS malfunction) arose.

aburkefl@gmail. Posted at 2015-6-22 20:45
"Gee, Mr. Doe. We're sure sorry about your Phantom that ran amok, seriously injuring several peopl ...

They're now called Airbus Helicopter. Aérospatiale is the French group that combined the aerospace companies of France.

Worked there many years ago.

Here's the reference to the suit I believe you were alluding to. https://www.law.cornell.edu/supremecourt/text/482/522

scott7161@gmail Posted at 2015-6-22 15:57
I've noticed the same thing even with firmware 1.1.9 on my Phantom 3 Pro.  At the moment I'm super c ...

Hi Scott
could you be just a bit more specific,  The problem that I have also was present with the previous versions,  Some of the very serious ill affects caused by upgrades, seems not to have affected mine except the vps now having to be turned off,  because of the weird ground level affects, it did try to take off on it own once though.

I seem to see slip without the altimeter saying as such, after a flight, my altimeter is still reading 30 to 50 ft altitude,    I think I can see the slip when ever you goose it or stop, the longer I fly, and the higher I fly ,  the worse it is,  One moment it seems to appear a software problem, hopefully    and the next
moment I think I am going to have to send it in.    detail your observations a bit for us if you could thanks

gregg1r Posted at 2015-6-22 20:33
Sploodge,
  Determining root cause of an issue is more difficult than just purchasing two or thre ...

I fully understand the elements involved in diagnosing possible issues and the variables..

Most of the people having issues start with "I did everything correctly then....". Granted testing the hardware on a scientific level is not possible but all the others bar climate/location is if "I did everything correctly then...." is true. This is what I doubt very much with some of the issues.
In the end this will likely just be a firmware fix altering some values to take into account some sensor tolerance issues some peoples Phantoms are experiencing.

My P2 has travelled with me to 8 countries on 5 continents and never had an issue other then a H3-3D that needed calibrating and I had no laptop . I have never seen such commotion from any other release like this P3 one and its due largely to so many new inexperienced people buying them and either crashing due to their own fault or some other random reason.

I don't accept the safety issues as if flown correctly and as per the law ( away from people ) then the risk is tiny. Yeah there have been reports on the news but almost all were due the pilot flying in places they really should not.. I sure DJI and other manufacturers have this angle covered
I had a recall on my old BMW Hybrid ( sold it as didn't really suit me really ) a few months back for a software issue related to braking. I was told no safety issue there Maybe Phillip programmed my brakes

I am here to help people who quite obviously need help or they would not come here.. I have no magic answer to their issue so they can take advice and try things suggested, ignore it or send the Phantom back to DJI, makes n differences to me. I would really appreciate it though if they were not dicks about it taking their frustrations out on someone who is trying to help them. It wont fix their issue but rather get them ignored.

mtnmaddman Posted at 2015-6-22 13:42
Man, finally that is exactly the same thing mine is doing, exactly!  I have posted at least a hundre ...

i have the same issues.  i got a repair request in the mail and am ready to send out for repair but i also hope it is a software issue still.  my problems started a few weeks ago and been upadating whenever a update came out.   i will make a  video of a auto takeof and post it online this evening. please let me know what u think of my video.

is there software that will let me read out the barometer and compare against a baseline or something like that?or enable a logging and testfly so i can send it in for examination by dji tech support?

and yes i too say it's dangarous, anyone thinking its not should mabe be grounded.

pete

Since I've had mine, it tends to lose a little altitude when hovering fairly close to the ground.  It also drifts a little bit, more of a back and forth movement.  This is with VPS off and on GPS with 13-15 satellites.

My iPad automatically updated the pilot app and now the altitude loss seems to be less....but only one flight so far.  Still on FW 1.1.9.

dronemastayoga Posted at 2015-6-30 02:43
i have the same issues.  i got a repair request in the mail and am ready to send out for repair but ...

I am not sure but the flight record in your pad should show the weird altitude readings,  I was going to make a real quick flight record of the problem, and open a ticket to dji.  If i were to take off fly up quickly to a few hundred feet and come right back down, I will be showing 30 to 40 ft alt. back on the  ground.  Sick of the problems,  I opened a dispute on pay pal, if the update does not come soon or doesn't work I may send back and try to buy a later version.  This thing looks more and more like a hardware problem on a totally integrated board, Who knows Dji is not about to say anything,  I am waiting for next update.

KaosImagery Posted at 2015-6-30 05:24
Since I've had mine, it tends to lose a little altitude when hovering fairly close to the ground.  I ...

Have you flown up to several hundred feet, and then after the flight "before you shut the motors off,"
looked at the altimeter reading, its easy not to notice, mine always reads 36 ft. when I bring it back to my eye level,  It is very consistent . being off by 30 ft after a flight . seems like others are off similarly.

Barometer is for height (vertical) distance, not horizontal hold.  Absolute/relative altitude, low pressure systems, high humidity, high temperatures affect barometer.

DaGunn Posted at 2015-6-30 08:40
Barometer is for height (vertical) distance, not horizontal hold.

Exactly my thoughts. The barometer that measures the altitude seems very accurate to me.

I would imagine a drift-issue as described would be GPS-related?

Personally, I can go straight up, and come straight down and come down within a meter or so.

I believe that if it drifts while moving (not going straight) it is likely to be a compass issue.

As the common universal solution seems to be; FULL IMU-calibration. On a completely flat surface, let the Phantom sit still and do its thing. Start out with a cool/cold Phantom (not least to avoid the "IMU Warming up" waiting time). I actually put mine in the fridge (for 15 mins or so)  before IMU-calibration. Less will do. If you can cool down a room or a car with aircondition and leave the Phantom in there, that might do the trick.

Good luck.

PS: Still on Firmware 1.1.9

The Phantom uses a barometric altimeter.

These types of altimeters are sensitive to temperature. As the temperature rises (as the aircraft warms up) or as the temperature falls (as you get higher and the temperature around the aircraft falls), the altimeter will give different readings. So wind, altitude, or anything that affects temperature might also affect the reading. Sudden changes in air pressure (owing to wind, cold fronts, etc.) will cause your altimeter readings to vary wildly.

Today I noticed that my Phantom will actually sink if left a few metres above ground. I tested this by aligning the Phantom with a tree in the distance in my line of sight, and noted that sometimes when it dropped, it wouldn't climb back up to where it was before. Within a minute or two, it was just half a metre from the ground after sinking about 2 metres. However, in an earlier flight, I noticed that it remained pretty much accurate throughout the flight (since I happened to notice negative altitude readings as I began to fly it down a hill I was on, so it must have been more or less accurate). This particular flight was the second of two consecutive flights, which leads me to believe that it is based more on the aircraft warming up than anything else.

An altimeter of this kind is not really supposed to give absolute readings - instead, its primary function is to detect relative changes, so the Phantom can tell if its going up or down. As such, you can expect the reading to be inaccurate after flying around for a while, as the error begins to build up and the environment changes. You can also expect some drift in altitude over time. It is kind of annoying that you have to keep an eye on your Phantom if you want to hover in place for a while, but if you're close enough to the ground for it to matter or be unsafe, then perhaps consider that the Phantom was made for aerial footage - get high up there and a few feet here or there won't make any difference ;)

Make sure VPS is off. If it shows off. Turn it on then back off. Do IMU calibration also. I had all of these same problems and was able to seem to fix by turning VPS on and off several times until everything got better. Also.....stay way above trees!!!!!!

FantomDK Posted at 2015-6-30 08:48
Exactly my thoughts. The barometer that measures the altitude seems very accurate to me.

I would  ...

My error readings are with the vertical (altitude) readings,  not a drift or horizontal telemetry error,  Some have stated that they are having that problem, that there home point is drifting.  That would be, in my opinion a gps situation.  

The problem discussion I think was originally directed at the barometric sensor,  which is of course measures vertical variations in the telemetry.  I would assume that inaccurate altitude readings would be a barometric sensor reading error, which I certainly hope are software related. and not a hardware problem.

When I first got my p3p about a month and a half ago now,  I would occasionally, get a barometric sensor error message, if the p3 sat to long on, before I started the motors,  I would have to cycle the bird to get the motors to light.  

Since the updates everything changed, I think in my case things got worse.  I do not receive the barometric sensor error message., but now I get an unidentified error 402, motors wont start until I cycle aircraft, everything is then fine and the motors will light.  So I don't know.

My horizontal telemetry is incredibly accurate , Like to the foot.  but my altitude reading upon return from flight is way off , according to the altimeter reading (before you shut off motors)  my craft is still 30 to 40 ft off of the ground.  Man I wish it was fixed.

Oliver Posted at 2015-6-30 09:05
The Phantom uses a barometric altimeter.

These types of altimeters are sensitive to temperature. As ...

Hi Oliver
I don't think we should have to just accept the problems with the sinking or the altimeter reading problems, or anything else, I have a multitude of naggy little problems I lost my lightbridge distance in the last upgrade , laggy video,  inaccurate hover, and inaccurate  altitude readings, on and on.

My Phantom v2+  has none of these problems,  never has, it has been absolutely rock solid, hundred of flights.  I did not see this coming at all.  Please don't just accept the problems you described that you are experiencing, they should not be there, I certainly hope that a software upgrade corrects these problems.

This is incredible technology, I love this thing, but I demand that it works as it is supposed to, as advertised, what I expected, and what I paid for, or it is going back.  I will wait until the next software upgrade comes out, if all of these problems are still there, it is going back, I will wait for the p4, maybe buy an inspire  or hope that a competitor can develop something better.

update:  here the video i made last night.  3x take of and dropping down by itself.


this dropping also happens sometime when comming down and having been flying for some time.
i had this drome hanging in stromg wind a few weeks (and firmware updates) ago, when wind was so strong i had to step  5 feet away from it to feel the downwind.
right under the drone was no wind because it was such a strong wind.   the drone was steady as a rock and did not move up down or left - right.
Now, even in s soft breeze or no wind, it moves.
is there a way to go back 4 firmwares to exclude the updates as a cause?

Looks like you have the VPS issue. I always have to give the automated take-off a helping hand as it will drop every time - I have VPS turned off for now too because of the latest firmware issue.

I see you have the IMU warm-up wait also...do a cold IMU calibration to fix that...  

DaGunn Posted at 2015-6-30 08:40
Barometer is for height (vertical) distance, not horizontal hold.  Absolute/relative altitude, low p ...

Hi DaGunn
I would think that the barometer resets the value that is fed to the telemetry system each time you light the p3 up, as it establishes a home point, and when you turn the motors on and off.  You can observe this on the app.

I would also think that the reading during the short period of each flight should remain somewhat consistent to only the altitude change that is occurring during the flight.  The only other factor that might have significant change in that short period of time I would think could be temperature.  My altitude reading errors seem to be very consistent, no matter the circumstances.

My P2V+ is absolutely solid and accurate in regards to telemetry issues.  The P3 is supposed to be a step forward in technology and product development and reliability.  At this point it certainly is not,  If Dji would offer a camera upgrade to my p2v+  I would consider going back to it.

We must not accept that it is alright if one of the main components of safe flight is not functioning correctly,  I would not think that a pilot would take his craft out for a flight with altimeter and other telemetry instrument reading errors.

dronemastayoga Posted at 2015-6-30 15:30
update:  here the video i made last night.  3x take of and dropping down by itself.

Looks pretty stable to me?

Question is; would it continue to drop if you didn't push it up after the auto-take-off? I doubt it - but understand why you'd prefer not to test it...

Why do auto-take-off anyway? Must admit, I've never done an auto-take-off, so I don't know how it acts.

Try with VPS turn off I guess.

Regarding the long warm-up. That is not related (to my knowledge) to the new firmware. It is related to the temperature of the aircraft when you did the Full IMU calibration. It has to get up to the temperature it was. This is why many use a trick - to cool down the P3 before doing the Full IMU calibration. Some call it "the refrigerator-trick" - some think thats going too far. But I've done that - put my P3Pro in the fridge for 15ish minutes (it was room temperature when I started - it had not just flown) and then took it out, put on a flat floor and did the full IMU calibration immediately. That seems (to me) to make the IMU work better - and there is zero waiting time for "IMU to warm up"! Meaning; more flying-time

You could cool your P3 by putting in a cold room, maybe a room you can chill with aircon - or put it in a car where the aircon has made it cool. Some just put it outside in the evening if it is cool outside. But the fridge has worked for me (do at your own risk naturally).

Good luck

As a pilot of an ultralight (U.S. Designation) having the altimeter drift by 30ft is normal. In my aircraft I have 2. The (better) electronic  one will often be 30 ft off after it warms up. The p3p I fly gets considerably  warmer after a short flight. I'm not surprised the reading changes. To be honest I rarely pay attention to that reading, when it's near the ground,I can see it.I've notice a difference between altimeters in different aircraft. Obviously aircraft need to alter the altimeter pressure setting to allow for weather and different airfield altitudes. But I have seen a difference of 60 ft when different aircraft are set to the same pressure altitude. They are all able to be re calibrated, the phantoms do it every start up but they are not of the same quality as full size aircraft. Why bother what it reads? It will give you a close enough reading at high levels, closer the floor who cares, it's only a number, you're the pilot.
RayH

ScottyT Posted at 2015-6-30 16:14
Looks like you have the VPS issue. I always have to give the automated take-off a helping hand as it ...

thanks. i did the cold recalibration yesterday, going out now to test it.
u remarked it looks i have the vps issue, could u tell by the dropping or other behavior u saw on the video.

peter

rhutty@gmail.co Posted at 2015-7-1 14:13
As a pilot of an ultralight (U.S. Designation) having the altimeter drift by 30ft is normal. In my a ...

Thank you for that infomation

dronemastayoga Posted at 2015-7-1 16:31
thanks. i did the cold recalibration yesterday, going out now to test it.
u remarked it looks i ha ...

How'd it go?

Just going by the video. I always have that issue during takeoff now - if you left it it would probably go back up in time, but not worth trying. After ascending to a height then coming back down I find it is stable again a few feet off the ground.

I recommend disabling VPS if you have the latest buggy firmware. It only really helps indoors, and I guess for doing smooth automated landings. An RTH landing without VPS can be pretty rough on the gear.  

mtnmaddman Posted at 2015-6-22 13:42
Man, finally that is exactly the same thing mine is doing, exactly!  I have posted at least a hundre ...

I have also problem with alttitude reading. Before take off it is 0 meter, but after 15 min flight, alttitude is 15 meter too high on ground level (home point). On the picture, my P3A was on the ground, but still alt was 20 meter...
-P3A
-FW 1.3.20

[img]https://forum11.djicdn.com/data/attachment/forum/%5BURL=http://s1032.photobucket.com/user/FCPhantom/media/Screenshot_2015-08-15_18-45-13_zps5jm8unlc.png.html%5D[/URL][/img]

jorski@netti.fi Posted at 2015-8-23 04:52
I have also problem with alttitude reading. Before take off it is 0 meter, but after 15 min flight, ...

Yeah 60 ft off that is ridiculous I sold my p3p because of this and the shell cracking issue, got an inspire had it three weeks did upgrade, it has now been sent off to dji with an esc error it wont even fly now, I have a p3 advanced also and a p2 these are my last dji products.

According to DJI, the normal barometer error is 2 cm / second.

If the flight time is 25min = 1500s, then the error can accumulate approximately to 30m = 2cm/s * 1500s

Since the p3 has an advanced gps system, I don't understand why it isn't used to periodically recalibrate the barometer reading

sploodge Posted at 2015-6-22 15:26
Have whatever opinion you want Phillip. I am only trying to help people but clearly you don't need ...

he must have hit a funny bone, dont see you get rattled too often. but you are absolutely right.
imo you are one of the ones who does try to help, a little on the dry side but, this is mostly a forum to help and be helped. keep up the good work and thanks

Same issue here... it´s a backstep... I get error about 10 meters. Vps says 2 meter... H says 15 meters.

automatic landing works fine ( since it uses VPS, I think), but I didn´t test with VPS OFF... maybe it read only gps altitude, and then it can down at 15 meters "thinking" it´s on air, and it´s on floor.

I don´t understand this issue... previous firmware was working fine to me (Heigh read), phantom 2 vision + worked perfectly too.

regards

Error is ALWAYS + (never -), always 10-12 meters (more)
regards

Hello to all. I m new. Yesterday third day with my p3p, I almost crash it. I didn't realized it. Only when I saw the video. I took my phantom in a remote area. I lifted it 120meters above me and fly it straight for a 1000meters. All the way the telemetry was spot on 120m height. The problem was that the terrain under was full elevation... And in the video I saw last meters almost touch small trees. I was lucky ! Few more meters and  crash for sure. Why my telemetry show me the 120 m reading all the time ??? Of course above me in the landing phase the telemetry works as Phantom lands. First day I update to the latest firmware both rc and craft. Thank you for your time.